Susceptibility weighted imaging
Citation, DOI & article data
Susceptibility weighted imaging (SWI) is an MRI sequence that is particularly sensitive to compounds which distort the local magnetic field and as such make it useful in detecting blood products, calcium, etc.
SWI is a 3D high-spatial-resolution fully velocity corrected gradient-echo MRI sequence 1-3. Unlike most other conventional sequences, SWI takes advantage of the effect on phase as well as magnitude 4.
Compounds that have paramagnetic, diamagnetic, and ferromagnetic properties all interact with the local magnetic field distorting it and thus altering the phase of local tissue which, in turn, results in a change of signal 2.
Paramagnetic compounds include deoxyhemoglobin, ferritin and hemosiderin 1.
Diamagnetic compounds include bone minerals and dystrophic calcifications 1.
Following the acquisition, post-processing takes place which includes a high-pass filter, to remove background inhomogeneity of the magnetic field, and the application of a phase map to accentuate the directly observed signal loss 2,4.
Typically the images presented are:
- filtered phase
- SWI (combined post-processed magnitude and phase)
Often a fourth set of images is provided, minimum intensity projection (minIP) which is just a thick slab of the conventional SWI images and is better able to demonstrate venous anatomy.
The most common use of SWI is for the identification of small amounts of hemorrhage/blood products or calcium, both of which may be inapparent on other MRI sequences.
They are also well suited to assess veins as deoxyhemoglobin results in both a loss in magnitude and a shift in phase 4.
Distinguishing between calcification (made up primarily of calcium phosphate, but also contain very small amounts of copper (Cu), manganese (Mn), zinc (Zn), magnesium (Mg), and iron (Fe)) 3 and blood products is not possible on the post-processed SWI images as both demonstrate signal drop out and blooming.
The filtered phase images are, however, able to (in principle) distinguish between the two as diamagnetic and paramagnetic compounds will affect phase differently (i.e. veins/hemorrhage and calcification will appear of opposite signal intensity) 3.
This is, however, not without its own complications as whether a lesion appears black or white on phase imaging depends on numerous factors:
- handedness of the system
- how images are presented (e.g. greyscale inversion)
- size and degree to which a lesion causes a phase shift (aliasing)
These are discussed below in the "pitfalls" section.
Despite the aforementioned difficulties, there are a number of tricks that enable you to correctly interpret whether a lesion is composed of dystrophic calcification or blood products on phase images in most cases.
Firstly, you will probably become familiar with your scanners and know if they are right or left-handed systems. This does not, however, prevent the images being greyscale-inverted. Thus the first step is to find a reliable internal control to establish how paramagnetic (e.g. blood products) and diamagnetic (e.g. calcification) will appear. Generally, the internal cerebral veins are readily identified. If the patient has pineal or choroid calcification this can also be helpful, however, due to aliasing (see below) this can be more challenging.
Once you have established this, then look at the lesion. If you are lucky it will appear uniformly either black or white, in which case you are finished. If it is the same as veins it is paramagnetic and therefore contains blood products. If it is the opposite, then it will be diamagnetic and therefore most likely dystrophic calcification.
In many instances, however, the lesion will have both black and white components due to aliasing. In this instance looking at the superior-most and inferior-most images is the key as they will be imaging not the lesion itself, but the magnetic dipole that matches the direction of phase shift of the central lesion, without the effect of aliasing. It can be helpful to reconstruct your images in the coronal plane to accentuate this effect 6.
Unfortunately, how phase information is shown is not uniform and varies from vendor to vendor and sequence 5. This is referred to as 'handedness' depending on the direction a positive change in phase is shown. If clockwise, this is referred to as a left-handed system because when you curl your left hand into a fist the fingers curl in a clockwise direction. This will in turn affect whether paramagnetic and diamagnetic materials will appear as white or black.
Windowing and greyscale inversion
As if the variability in handedness was not sufficiently annoying, how phase images are presented is also variable. And, to make matters worse, if filtered phase images are grey-scale inverted they will look very similar. This can cause confusion of mistakes when interpreting phase images.
A simple step to make sure you avoid this is to first check the signal intensity of a known structure e.g. the internal cerebral veins. This will tell you what paramagnetic compounds should look like.
If you want to, you can also invert the image if need be so that you always look a phase images the same (e.g. veins appearing dark).
Although filtered phase images are probably more sensitive to minimal amounts of calcium than CT 3, they perform poorly and can be confusing when larger amounts of calcification or hemosiderin are present. When the ﬁeld is large enough that the phase exceeds π (pi) radians, it will alias to -π radians and will now appear to be dark rather than bright 3. The net effect is that large regions of calcifications can have areas that appear dark, or be surrounded by dark regions. The converse is also seen in large areas of profound hemosiderin staining.
Patients that are receiving substantial supplemental oxygen (e.g. intubation) may have very little deoxyhemoglobin in venous blood, resulting in limited visualization of venous structures. This is due to oxygen in hemoglobin shielding the iron from having their usual paramagnetic effect 4. The same effect is taken advantage of to perform blood oxygen level dependent (BOLD) functional MRI.
- 1. Schweser F, Deistung A, Lehr BW et-al. Differentiation between diamagnetic and paramagnetic cerebral lesions based on magnetic susceptibility mapping. Med Phys. 2010;37 (10): 5165-78. - Pubmed citation
- 2. Tong KA, Ashwal S, Obenaus A et-al. Susceptibility-weighted MR imaging: a review of clinical applications in children. AJNR Am J Neuroradiol. 2008;29 (1): 9-17. doi:10.3174/ajnr.A0786 - Pubmed citation
- 3. Wu Z, Mittal S, Kish K et-al. Identification of calcification with MRI using susceptibility-weighted imaging: a case study. J Magn Reson Imaging. 2009;29 (1): 177-82. doi:10.1002/jmri.21617 - Free text at pubmed - Pubmed citation
- 4. Barnes S & Haacke E. Susceptibility-Weighted Imaging: Clinical Angiographic Applications. Magn Reson Imaging Clin N Am. 2009;17(1):47-61. doi:10.1016/j.mric.2008.12.002
- 5. Haller S, Haacke E, Thurnher M, Barkhof F. Susceptibility-Weighted Imaging: Technical Essentials and Clinical Neurologic Applications. Radiology. 2021;299(1):3-26. doi:10.1148/radiol.2021203071 - Pubmed
- 6. Weng C, Jeng Y, Li Y, Chen C, Chen D. Black Dipole or White Dipole: Using Susceptibility Phase Imaging to Differentiate Cerebral Microbleeds from Intracranial Calcifications. AJNR Am J Neuroradiol. 2020;41(8):1405-13. doi:10.3174/ajnr.a6636 - Pubmed